Affiliation:
1. Vellore Institute of Technology, India
Abstract
Cancer patient survival is an integral part of the healthcare sector, with researchers offering clinicians snipping information as they consider treatment options that have a big impact on patients' lifestyle decisions. Evidently, no study of predicting survival or death from breast cancer using XGBoost method has been attempted. The goal of this project is to design a prediction system that can forecast both the survival rate and the death rate from breast cancer at an early stage by examining the most limited collection of clinical dataset variables. The potential of the XGBoost method is determined using mainly age at the time of diagnosis along with other clinical characteristics, assessed for its relative significance. The project findings show that the XGBoost model fits the testing dataset more effectively, with outcome of accuracy 82.72% approximately.